© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7896
    Prev
    Next
    INDEX
    Explanations

    adjectives describing states or qualities

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on words conveying aversion or fear (negative sentiment terms like “loathed” or “gun-shy”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ע
    0.62
    Czas
    0.52
    எ
    0.51
    א
    0.51
    יש
    0.50
    Another
    0.50
    என்
    0.50
    #!/
    0.47
    Een
    0.47
    ❮
    0.47
    POSITIVE LOGITS
     ferrite
    0.64
     sparsebundle
    0.64
     Foundry
    0.62
     electrolytic
    0.61
     transcoding
    0.61
     Admiralty
    0.60
     microglia
    0.59
     lackluster
    0.59
     WebView
    0.57
     einzigen
    0.57
    Activations Density 0.000%

    No Known Activations